In this I decided to compare the difference of blacks on welfare and whites on welfare. While it may appear that there are more whites total on welfare, when broken down by race, there is a higher percentile of blacks on welfare as they have less total population than whites. If there was the same amount of blacks in the United States as there were whites, then whites would have the highest total people on welfare while blacks would have the lowest.
(So you're making a mistake here, you're overestimating what your'e analyses actually say. In a sense, that's not bad--it's really cool you're mind is pushing forward into really interesting and important substantive questions. But in another sense, you must not be perfectly clear on what these techniques are doing to the data! More on this below....)
summary(nes$blacks)
Min.
1st Qu.
Median
Mean
3rd Qu.
Max.
NA's
0.00
50.00
70.00
72.04
85.00
100.00
267
sd(nes$blacks, na.rm=TRUE)
[1] 20.64901
summary(nes$whiteppl)
Min.
1st Qu.
Median
Mean
3ed Qu
Max
NA's
0.00
50.00
70.00
73.35
85.00
100.0
270
sd(nes$whiteppl, na.rm=TRUE)
[1] 19.48738
summary(nes$welfareppl)
Min.
1st Qu
Median
Mean
3rd Qu.
Max
NA's
0.00
40.00
50.00
56.73
70.00
100.00
276
sd(nes$welfareppl, na.rm=TRUE)
[1] 21.43782
Boxplots:
Key:
Blue=Blacks
Orange=Whites
Green= People receiving welfare
Or if you prefer a prettier version:
Whites and Welfare
Blacks and Welfare
I believe one could use this to say that there are more whites receiving welfare than blacks or the other way around.
So, you're execution of analyses was excellent, but we have to be careful! You're making claims about the world that are WAY bigger and harder to demonstrate than what you're data is doing. Like I said, that's kind of great! Don't lose your ideas; hold them, because you're going to learn how to test these kinds of hypotheses you're articulating here! But go slower. The tools you use here only let you articulate basic descriptive summaries of the variables and the correlation says a little bit about how two variables may be associated. But the data doesn't really provide evidence for the claims you're making--you might be right, but you need to be a little bit more creative (with tools you don't have yet!) to substantiate these claims! I really really like your eagerness to make big, interesting claims, though. Right on!)
(So you're making a mistake here, you're overestimating what your'e analyses actually say. In a sense, that's not bad--it's really cool you're mind is pushing forward into really interesting and important substantive questions. But in another sense, you must not be perfectly clear on what these techniques are doing to the data! More on this below....)
summary(nes$blacks)
[1] 20.64901
summary(nes$whiteppl)
[1] 19.48738
summary(nes$welfareppl)
[1] 21.43782
Boxplots:
Key:
Blue=Blacks
Orange=Whites
Green= People receiving welfare
Or if you prefer a prettier version:
Whites and Welfare
Blacks and Welfare
I believe one could use this to say that there are more whites receiving welfare than blacks or the other way around.
So, you're execution of analyses was excellent, but we have to be careful! You're making claims about the world that are WAY bigger and harder to demonstrate than what you're data is doing. Like I said, that's kind of great! Don't lose your ideas; hold them, because you're going to learn how to test these kinds of hypotheses you're articulating here! But go slower. The tools you use here only let you articulate basic descriptive summaries of the variables and the correlation says a little bit about how two variables may be associated. But the data doesn't really provide evidence for the claims you're making--you might be right, but you need to be a little bit more creative (with tools you don't have yet!) to substantiate these claims! I really really like your eagerness to make big, interesting claims, though. Right on!)